Bias in the jury selection process
When selecting a jury, both the defense and the prosecution are allowed to strike potential jurors from the pool. While the potential jurors provide answers to a questionnaire, what kind of role might race play in their selection or rejection?
logistic regression feature selection odds ratio p values race
Readings and links
- How did Curtis Flowers end up with a nearly all-white jury?
- In the Dark, Season Two
- GitHub repo with code and data
- Whitepaper for the analysis
Summary
Building on our experience with logistic regression, can we find bias in the jury selection process? Our feature encoding is pretty easy, but there are a lot of things we'll be controlling for.
Notebooks, Assignments, and Walkthroughs
Combining and cleaning the initial dataset
Cleaning the dataset
Picking what matters and what doesn't in a regression
How to use the statistical measure "p-value" to pick which features to include in your regression model. Also known as "feature selection."
Analyzing data using statsmodels formulas
Perform some cleaning and the main logistic regression using the formula method in the statsmodels package.
Alternative techniques with statsmodels formulas
Using casting-as-categorical in statsmodels formulas to approach this in another way. Not the same as